Psychological Effects of a Synchronously Reliant Agent on Human Beings

Abstract

The ability of human symbiosis robots to communicate is indispensable for their coexistence with humans, so studies on the interaction between humans and robots are important. In this paper, we propose amodel robot self-sufficiency system that empathizes with human emotions, a model in which we apply the urge system to an autonomous system of emotions. We carry out simulation experiments on this model and verify the psychological interaction between the software robot and its users.

Psychological effects of educational-support robots using an emotional expression model

“…Second, the robot expressed autonomous emotions. Previous studies have indicated that a sympathetic agent enhances learning motivation [8] and that an agent using a model of emotional expressions facilitates more effective interactions with humans than the one expressing random emotions [5]. A similar effect was observed during collaborative learning between a learner and the robot.…”

“…3), which has previously facilitated human-robot interactions. Moreover, this earlier report suggests that the agent, which uses Russell's circumplex model of affect, can prompt more effective interactions with humans than an agent using a conventional model [5]. In addition to the proposed model, this previously described conventional model, which utilizes one coordinate point to express emotions, is also evaluated for comparison.…”

“…This model has proved beneficial for agent-human interactions. An agent reported that using this model has facilitated a more effective interaction with humans than the one expressing random emotions [5]. Moreover, an agent presenting a positive expression has achieved collaborative learning with humans more effectively than the one showing negative expression and no emotion [6].…”

A proposal of model of emotional expressions for robot learning with human

“…1), which has previously been used to facilitate human-robot interactions. Moreover, an earlier report suggests that an screen agent that uses Russell's circumplex model of affect can prompt more effective interactions with humans as compared with an agent using a conventional model [6]. In this study, the pleasure-displeasure and arousing-sleep axes correspond to a student's number of correct answers and their answer times, respectively.…”

“…This model has proven beneficial for sustained interest in agent-human interactions. It has been reported that screen agents using this model facilitate more effective interactions with humans than do those that only express random emotions [6]. Moreover, it has been shown that a screen agent expressing positive emotions achieves more effective collaborative learning than does a screen agent expressing either negative or no emotions [7].…”